where ? k is the vector of h coefficients corresponding to the k th occupation group. The total number of parameters to be estimated is h(N – 1), since coefficients for each element of x are determined only up to an arbitrary normalization. Thus, the estimated coefficients for each occupation group do not represent marginal probabilities, but indicate the change in the log of odds of being in that occupation instead of in unskilled occupations.
Because of the ranks the fresh coefficients out of certain varying by the proportions, we are able to comprehend the relative impact of these adjustable for the probabilities of being in certain employment
The estimates of one’s multinomial logit model mean that discover differential entry to job predicated on gender (find upper panel out of Table 8). The brand new coefficient towards the feminine dummy is a lot distinctive from no in five community groups and the indication may vary all over job. Ceteris paribus, the possibilities of becoming a manager or plant and server operator is leaner (the newest coefficient with the female dummy is actually bad and you may high) additionally the likelihood of getting a specialist, professional and user top-notch or a clerical staff is higher (the new coefficient to your women dummy try confident and you can tall) for ladies versus men.
There isn’t any significant difference ranging from men and women regarding the probability of being involved because the provider and transformation experts or being inside activity and associated exchange
To measure the impact of differences in “tastes”, hiring prejudices, or other unobservable factors on the occupational distribution, we estimate a separate model of occupational attainment for men (see lower panel of Table 8). Employing these estimates, we obtain the predicted distribution for each sex as follows: substitute the sample data for each sex into the estimated model, producing for each individual a vector of predicted probabilities of belonging to each of the eight occupation groups, and calculate the mean of the predicted probabilities for each occupation after summing over observations. For men this estimation yields a predicted distribution which is identical to their actual sample distribution, i.e., M p = M a . The difference in the predicted distributions of men and women (M a –W p ), is the ‘explained’ component due to difference in characteristics, and the residual difference, (W p –W a ), is the ‘unexplained’ component due to differences in tastes, hiring prejudices, or other unobservable amerikkalaiset vs. venГ¤lГ¤iset naiset factors.
The outcome of one’s decomposition take action was showed when you look at the higher panel regarding Table 9. We appeal mainly on the decompositions to your large differences in the real sample distributions. The newest do it shows that if there is certainly zero differential access to occupation because of the gender, the latest ratio of women engaged just like the benefits, auto mechanics and you will member advantages, and you may clerks would be much smaller than noticed, while the ratio of women interested once the managers and you will bush operators could be much more greater than noticed (column 5). The residual “unexplained” part makes up around 73-78 percent of your own seen difference between people within the the new dimensions engaged in elite and clerical occupations, and you will 97 per cent in the example of bush and you can server providers (column 7). Because of the nature of them employment, which probably shows “supply-side” variations in choices of females for light-neckband jobs. To own professionals and you may elderly officials, the rest of the “unexplained” differences (0.0269) is simply more than the fresh new noticed difference (0.0198). In the event that you will find zero differential access to career of the gender, the latest ratio within the managerial ranking could be large for females than for men. It is hard so you’re able to ascribe so it selecting to “supply-side” difference between choice.